FORECASTING OF TROPOSPHERIC OZONE CONCENTRATIONS USING ARTIFICIAL NEURAL NETWORKS IN THE VICINITY OF A THERMAL POWER PROJECTS. Anand Kumar Varma and K.R. Manjula
Nitrogen oxides (NOx) and Volatile organic compounds (VOCs) result in the formation of ozone in the troposphere as an outcome of a chain of intricate reactions between them and are thus the precursors of ozone. Thermal power plants are one of the major sources of these precursors. The present investigation deals with identification of an important model, specifically with the function of an Artificial neural network (ANN) to simulate the forecasting of surface ozone in the vicinity of a Thermal Power Plant. The ANN model was developed with neural applet version 4.3.8 to forecast diurnal ground level concentrations of ozone. Further ambient ozone concentrations were predicted using surface meteorological variables such as Temperature (Temp), cloud cover, rainfall, relative humidity (RH) along with the precursor NOx pollutant variable. Predicted and measured NOx values were used as predictors. Predicted NOx values were obtained from Industrial source complex short term version 3 (ISCST3) model. The results indicated that among all the meteorological inputs, temperature plays a significant role in the production of ozone and nitrogen oxide concentrations are directly related to the forecasted surface ozone concentrations. The predicted values of ozone from ANN model were found to be in good agreement with measured concentrations and thus validating the developed model. It was established that the measured concentrations and predicted concentrations by the model are reliable.
Enter your contact information below to receive full paper.